Imagine two pinball machines, the second of which has had exactly those pins removed that are never touched by the ball.

We now have two different machines, but it doesn't make a difference to the paths traced by the pinballs.

The counterfactual are different in each machine—the pinballs would behave differently if the first were to hit a "counterfactual" pin—but the physical activity of the two systems is, as it happens, identical.

Tim Maudlin (1989).

The Maudlin argument

Tim Maudlin argues this way:

"The notion that difference of counterfactual structure implies difference in physical process does not fare well when confronted with cases.

Maudlin presents this case:

"The deterministic pin-ball example. "Consider, for example, a deterministic pin-ball game. If we set the ball off at a specified place and with a specified velocity, it will always trace exactly the same path through the board. The only physical processes involved are those associated with the motion of the ball down the board and the interactions of it with the pins that it encounters. If we now remove a few pins, pins which the ball never touches on this path, pins which perhaps it never even comes near, the ball will continue to retrace exactly the same path in exactly the same fashion. The physicist's explanation of why it traces just that path will remain exactly the same. No motion or energy transfer or change of state occurs anywhere outside the path of the ball. From a physical point of view, the processes an actions that occur with or without the peripheral pins in place are identical, for there are no physical processes occurring outside the path. But the counterfactuals in the two cases are different. For had the ball been given a different initial push, a push that would have carried it into the region where the pins are removed, then its path would be different in the two cases. So, although the counterfactuals supervene on the entire physical state of the system, differences in counterfactuals about the evolution of the system need not imply differences in physical processes that are evolving at a time" (T. Maudlin, 1989, p. 424-5).

References

Maudlin, Tim. 1989. Computation and Consciousness. The Journal of Philosophy, vol. LXXXVI, no. 8, p. 407-432.